Understanding how organisms extract sensory information from their environment, encode it into electrical and chemical signals, then integrate and process these signals to produce behavioral responses conducive to survival, is a fundamental goal of neuroscience. A prerequisite for understanding how any given behavior is generated is knowledge of both the identities of the component neurons of the underlying neural circuit and the synaptic connectivity relationships among them. In this application it is proposed to adapt to Drosophila the second-generation GFP Reconstitution Across Synaptic Partners (GRASP) technique recently developed in the mouse. Unlike the original version of GRASP that had a high false-positive rate, this improved version of GRASP produces GFP signals exclusively at synaptic contact sites. Adaptation of this synapse-specific version of GRASP to Drosophila will tremendously enhance the ability to map the underlying neural circuits of any number of well-characterized Drosophila behaviors and thereby the understanding of how sensory information is translated into behavioral responses in this established model system with powerful molecular genetic advantages. In addition, it will be useful for eluciding the neural circuitry underlying Drosophila models of human medical conditions including neurodegenerative disease, addiction, and sleep.